A comparative study of indocyanine green instillation in inguinal node versus foot web space using da Vinci indocyanine green FireFly™ technology in identifying thoracic duct during robotic-assisted transthoracic oesophagectomy

Abstract

Introduction:

Chyle leak is a serious complication following oesophagectomy with incidence varies from 1% to 9%. Near infra-red fluorescence imaging of thoracic duct (TD) can provide real-time dynamic imaging during the surgery. In this study, we intend to compare indocyanine green (ICG) dye instillation through inguinal node with subcutaneous first web space instillation for visualisation of TD during robotic-assisted minimally invasive oesophagectomy (RAMIE) procedure.

Patients and Methods:

A prospective study of 50 patients underwent RAMIE with da Vinci X System. After general anaesthesia, patients were divided into inguinal node and foot first web space ICG instillation group. The former group had 1 ml of ICG dye instilled on bilateral inguinal nodes under ultrasound guidance and while the other group received 1 mL of ICG dye injected at bilateral foot first web space and then underwent surgery. TD was visualised using ICG FireFly™ fluorescence technology, first at the time of docking and subsequently for every 5 min until 60 min of instillation time and analysed.

Results:

Twenty-five patients were enrolled in each group. The mean docking time for thoracic phase was 13.76 ± 3.43 min. TD was visualised in 72% (18/25) of cases of first web space instillation group, whereas 100% in ultrasound guidance inguinal node instillation group. None of the patients had a chyle leak.

Conclusion:

ICG FireFly™ fluorescence technology for the identification of TD during oesophageal mobilisation is safe and effective and provides real-time dynamic visualisation with high accuracy in ultrasound-guided bilateral inguinal node instillation group. It is an effective method for the surgeons planning to negotiate their initial learning curve in RAMIE procedures.

INTRODUCTION

Thoracic duct (TD) is the tubular structure which runs at the posterior mediastinum of the thorax of 38–45 cm length. It originates from cisterna chyli at the L2 vertebral level and runs between the aorta and the azygous vein and anterior to the vertebral column in posterior mediastinum to drain into the left subclavian and internal jugular vein. Although TD ascends through the aortic hiatus of the diaphragm, it lies posterior to the oesophagus and crosses it at the T₇ level and crosses to the left side at around the T₅ vertebral level to drain into junction of the left subclavian and internal jugular vein.[1] Chyle leak is a serious complication can occur following oesophagectomy. Based on literature, the resultant chylothorax incidence varies from 1% to 9%.[2] Chylothorax is associated with considerable morbidity ranging from pneumonia and respiratory failure. Persisting indwelling drains could be due to high volumes of fluid lost from a TD injury, leading to hypovolaemia and malnutrition from the loss of emulsified fat, chylomicrons, fat-soluble vitamins, electrolytes, proteins and depletion of T-cells.[3] There is also a risk of sepsis around 24% secondary to lymphopaenia, pneumonia, respiratory and mortality reported as high as 75%.[4] It can be difficult to identify the site of TD chyle leak during re-operative procedures and several methods such as magnetic resonance-thoracic ductography and lymphoscintigraphy, using Tc⁹⁹ filtered sulphur colloid combined with single-photon emission computed tomography been used.[5,6] However, none of these methods provide dynamic real-time imaging in operation theatre. Best way to avoid chylothorax would be careful intra-operative dissection and restricting dissection close to oesophagus. Apart from this, other traditional techniques employed to visualise TD intraoperatively would be prior enteral administration of cream and methylene blue.[7,8] Due to fat and fascial covering over TD, these techniques will be useful once TD is dissected free from surrounding structures which further risks it from injury.[4,7] Near infra-red (NIR) fluorescence imaging of TD can provide real-time dynamic imaging during the surgery. Various sites such as subcutaneous tissue of groin, inguinal nodes, web space, mesenteric instillation or dye with heavy cream through enteral feed have been mentioned for the visualisation of the TD.[8,9] In this study, we intend to compare indocyanine green (ICG) dye instillation through inguinal node with subcutaneous first web space instillation for visualisation of TD during robotic-assisted minimally invasive oesophagectomy procedure at our centre.

PATIENTS AND METHODS

A prospective study of patients who underwent surgery for oesophageal malignancy cases at our Manipal Comprehensive Cancer Centre, Bengaluru, from January 2020 to July 2022 were considered for the study. Patients underwent McKeown oesophagectomy for carcinoma oesophagus using the da Vinci X Surgical System. The study was approved by the institutional ethics committee registered as ECR/34/Inst/KA/2013/RR-19. Inclusion criteria were patients above 18 years of age with biopsy-proven oesophageal cancers. Exclusion criteria were multiple previous abdominal surgeries not amenable for robotic approach, metastatic disease and patients with prior lung conditions unfit for oesophagectomy procedure. Patients’ demographic, medical history, pathology, tumour staging and procedural details with clinical outcomes were documented. Patients underwent prior upper gastrointestinal scopy and tumour location was confirmed. According to our institution protocol, all patients underwent positron emission tomography-computed tomography scan for disease evaluation and for distant metastasis. All the patients with squamous cell carcinoma of the oesophagus, underwent neoadjuvant chemoradiotherapy and the patients with adenocarcinoma oesophagus underwent neoadjuvant chemotherapy before surgery. Informed consent was taken. To detect a large effect of 0.8 between the two groups at 5% level of significance and 80% of power, a minimum of 25 subjects were required in the individual group.

Procedure steps

After induction under general anaesthesia with double-lumen endotracheal tube, diagnostic peritoneoscopy was done for the patients with adenocarcinoma of oesophagus to rule out the possibility of peritoneal deposits, and later, the patients were explained about the procedure and were allowed to choose between inguinal node and foot first web space ICG instillation group. Those with inguinal node group had 1 ml of ICG dye instilled on bilateral inguinal nodes under ultrasound guidance, and the second group had 1 ml of ICG dye injected at bilateral first web space of the foot in the subcutaneous plane and was massaged locally for 10 min to enhance the lymphatic flow. For inguinal node instillation, high-frequency ultrasound probe is used by the surgeon to identify pulsating femoral vessels in the inguinal, and thereafter, probe is moved proximally to focus on the subcutaneous plane where nodes are identified, fixed and instilled with 1 mL ICG dye on the either side [Figure 1]. Patients were later placed in the prone position and ports were inserted [Figure 2]; robotic cart was docked at the left shoulder. The TD was visualised using ICG FireFly™ fluorescence technology in da Vinci X surgical system, and time was noted according to the visualisation, first at the time of docking and subsequently for every 5 min until maximum 60 min of instillation time, following which dissection or the surgery was initiated regardless of visualisation of TD. Time zero was initiated during docking process excluding the time taken for dye instillation and time taken for the local massage.

Figure 1.

Indocyanine green instillation (a) bilateral foot first web space, (b) under ultrasound guidance at bilateral inguinal nodes, (c) ultrasound identification of inguinal node

Figure 2.

Port placements for robotic oesophagectomy under prone position

McKeown oesophagectomy was done in three phases: thoracic, abdominal and cervical phase. Thoracic phase was divided into infra-azygous dissection and mobilisation of oesophagus, dissection at the diaphragmatic crus, subcarinal dissection and ligation of azygous vein, supra-azygous dissection and clearance of recurrent laryngeal lymph nodal chain. TD was visualised with ICG FireFly™ fluorescence technology during the procedure. As per the routine surgery, later abdominal phase of lymph node dissection, stomach mobilisation and formation of gastric conduit was done. This was later continued with cervical phase of mobilisation and oesophagogastric anastomosis. Data were collected based on the docking time, operative time, blood loss, conversion to open, chyle leak and post-operative complications. Descriptive statistics was used to assess the data collected from the study using IBM SPSS Statistics for Windows, version 29 (IBM Corp., Armonk, NY, USA). Categorical variables were summarised as frequencies and proportions, whereas continuous data were expressed as mean ± standard deviation and median. Log-rank test was used to compare the results of the two groups.

RESULTS

Between January 2020 and July 2022, 50 patients were enrolled in the study with 25 patients were included in the first web space instillation and 25 patients in ultrasound-guided inguinal node instillation, respectively. The demographics and clinical data of the patients included in the study are summarised as in Table 1. The mean age of the carcinoma oesophagus patients in our study was 47.4 ± 6.7 years and 64% (32) of the study population were males with the mean body mass index of 23.21 ± 7.32 kg/m². 56% (28) of the cases were squamous cell carcinoma pathology. The mean docking time for both abdominal and thoracic phase of surgery was 33.20 ± 4.16 min, whereas the thoracic docking time was 19.76 ± 3.43 min. Hence, by default, our first recording time of the visibility was 16 min on an average, and from then, on every 5 min, ICG FireFly™ fluorescence technology was activated to check for the visualisation of the TD. Once visualisation of the TD was confirmed, oesophageal mobilisation with lymphadenectomy was done. After docking, maximum duration of 1 h was given for the visualisation of the TD at the infra-azygous portion of the oesophagus and the surgeon proceeded with the surgery if in case the fluorescence did not appear by the end of an hour. Accordingly, the time frame was adjusted while noting the total operative and thoracic phase operative time. 72% (18/25) cases of first web space instillation, TD was visualised, and in 28% (7/25) cases, it was not visualised, whereas 100% visualisation was confirmed in direct ultrasound guidance inguinal node instillation technique [Figure 3].

Table 1.

Demographic and clinical characteristics of the patients

Demographic and clinical data
Patient characteristics	n	Inguinal node group	Web space group
Age (years)	47.4±6.7	43.6±3.4	49.4±7.2
Gender (male: female)	32 and 18	17 and 8	15 and 10
BMI	23.21±5.32	22.56±3.24	25.1±2.38
Pathology
Squamous cell carcinoma	28	14	14
Adenocarcinoma	22	11	11
Mean docking time (min)	33.20±4.16	32.60±4.65	34.20±3.16
Thoracic docking time (min)	13.76±3.43	13.25±3.21	12.76±3.67
Mean operative time (min)	321.13±13.75	310.13±16.75	325.13±8.65
Thoracic phase operative time (min)	57.04±9.15	52.04±6.15	59.44±8.15
Mean blood loss (mL)	256.32±17.52	210.32±9.34	266.32±12.22
Median time of ICG in thoracic duct (min)		20	45
Conversion to open		Nil	Nil
Chyle leak		Nil	Nil

BMI: Body mass index, ICG: Indocyanine Green

Figure 3.

Results of first web space instillation versus ultrasound guided inguinal node instillation. CI: Confidence interval

The mean operative time was 321.13 ± 13.75 min, while for thoracic phase, it was 57.04 ± 9.15 min. The mean blood loss during the whole procedure was 256.32 ± 17.52 mL and none of the patients had a chyle leak, require conversion to open procedure or necessitated prophylactic ligation of TD [Table 1].

DISCUSSION

For oesophageal cancer, oesophagectomy is the only curative option. This procedure is associated with significant morbidity and mortality.[10] Minimally invasive approach for oesophagectomy has gained precedence in recent years, which has minimised surgical trauma and complication rate.[11,12] Over a past decade, robotic-assisted oesophagectomy in a prone position has been performed in our centre. Literature quotes mortality rate of 2.3% and morbidity rate of 25% within 30 days.[10,13] Although rare, chylothorax is a life-threatening complication of mediastinal lymphadenectomy. The incidence of chylothorax varies from 1% to 9%.[2] Conservative treatment of reduction of oral fatty food intake and parenteral nutrition has varied success rates of 3%–90% based on the disease status, site and extent of damage, especially in high-output fistulas.[14] Many of these patients tend to develop electrolyte imbalance, dehydration and malnutrition which may deteriorate post-operative course with mortality being 30% if not responding to conservative management.[15] Non-surgical therapeutic options such as lymphangiography with percutaneous embolisation techniques such as coils and liquid embolic agents for occlusion could be attempted with 90% efficacy based on the case-to-case basis.[16] Surgical management involves ligation of TD, chest tube placement, pleurodesis or pleuroperitoneal shunt along with conservative management with success rates ranging from 64% to 100%.[13,17] The best way to avoid chylothorax is to avoid damage to TD during the surgery either by dissecting close to oesophagus and mobilising it with poor quality mediastinal lymphadenectomy, intra-operative en masse ligation of TD with lymphadenectomy or by utilising better intra-operative visualisation techniques for oesophageal mobilisation.

Intra-operative en masse ligation or dissection of TD from fat and fascial covering over TD, further risks it from injury.[4,7] While resecting TD, Schurink et al. and Udagawa et al. demonstrated the presence of lymph nodes in the surrounding it but found no significant difference in survival in those group of patients.[18,19] Matsuda et al. later compared the outcomes after resection of the TD with the preservation of the TD showed a significantly higher yield of lymph nodes with recurrence-free survival.[20] Further, not only the TD resection is not only involved with increased operating time and blood loss but also is associated with post-operative complications.[21,22] Ohkura et al., in their study have mentioned that TD resection,has little or no benefit of reduction in the incidence of chylothorax. Lymphangiography techniques such as pre-operative oral administration of high-fat diet such as milk, heavy cream or olive oil with methylene blue a day before surgery and ICG with NIR fluorescence have gained prominence. The former technique is associated with the staining of the tissue on accidental injury to the TD and contamination of the surgical field, whereas the latter has gained prominence in the robotic surgery due to incorporation of ICG FireFly™ fluorescence technology in da Vinci systems. ICG NIR fluorescence has got excellent sensitivity and specificity by providing precise area of chyle leak with additional feature of real time visual confirmation of clipping of leak or its repair due to the pooling of the green fluorescence at the clip site.[23,24] ICG is a fluorescent molecule which emits photons when infrared light of wavelength 780 nm provokes it. The emitted photons are picked up by the dedicated optical systems and software in da Vinci systems to appear like fluorescence. Its intravascular half-life is about 2.4 min. When ICG dye is injected in the body, it binds to the lipoproteins until hepatic uptake and eventually excreted in the bile.[25] At the standard dose of 2 mg/kg in non-iodine allergic patients, it is non-toxic and hardly has any interactions. Together with high definition, 12×, three-dimensional stereoscopic vision, a stable and surgeon-guided camera and improved ergonomics, it provides real- time dynamic imaging in the tissues injected with the dye. For intra-operative TD visualisation by FireFly™ fluorescence technology, studies have quoted injecting ICG dye in the web space, inguinal node, popliteal node, groin skin, subperitoneal tissue, small bowel mesentery, enteral route and have quoted varied time frames for the TD visualisation. However, these are few studies which compared the efficacy using simpler techniques. In our study, we wanted to compare the simpler techniques such as bilateral first web space instillation with the bilateral inguinal node instillation under ultrasound guidance after inducing the patient under general anaesthesia so as to avoid the discomfort and anxiety of the patient of being subjected to the painful injection 10–12 h before the surgery. Based on the results of our study, we found that ultrasound-guided ICG instillation at the inguinal node was significantly superior to the bilateral first web space instillation. In the former group of patients, TD was visualised within 30 min of induction and subsequently helped in the dissection and reducing the learning curve for oesophageal mobilisation. While in the first web space instillation group, only 72% of patients TD were visualised at the end of 1 h [Figure 3]. Ultrasound-guided nodal ICG instillation process is easy to perform, not time consuming and does not require additional skill to perform as in popliteal nodal or direct mesenteric instillation.[8] TD is more prone for the injury during oesophageal mobilisation at the level azygous vein or during recurrent laryngeal nodal dissection. ICG instillation helped us to visualise TD at these levels where it lies in close proximation with recurrent laryngeal nerve, oesophagus and membranous part of tracheal at supra-azygous level [Figures 4 and 5].

Figure 4.

Intra-operative thoracic duct visualisation (a) below azygous vein dissection, (b) infra-azygous dissection, (c) at the level of diaphragmatic hiatus

Figure 5.

(a-c) Thoracic duct visualisation at the level of supra-azygous dissection lying close to left recurrent laryngeal nerve

The findings of our study, i.e., visualisation of TD under 30 min is in agreement with various studies[23,24,26] and is in contrary to the findings of Vecchiato et al.[9] Matsutani et al. and Yang et al. found reasonable success in their technique of pre-operative groin subcutaneous ICG instillation for visualisation chyle leak.[27,28] Barnes et al. conducted a study, where visualisation of TD was compared with direct mesenteric instillation versus enteral route. They concluded that 70% TD were identified with mesenteric injection, while the enteral technique was ineffective.[8] These findings were similar to the findings of Kaburagi et al. where mesenteric route of ICG instillation technique was employed to detect chyle leak in patient who underwent upfront gastro-oesophageal junction cancer surgery.[29] Based on the findings and technique involved in our study, ultrasound-guided inguinal node instillation is safe, easily reproducible and less cumbersome than direct small bowel mesenteric, popliteal nodal or subperitoneal instillation and more reliable than web space or groin subcutaneous tissue instillation to provide real-time dynamic visualisation of TD. However, larger comparative studies can be designed comparing various modalities of ICG instillation for TD identification during oesophageal mobilisation.

CONCLUSION

ICG FireFly™ fluorescence technology for the identification of TD during oesophageal mobilisation is safe and effective and provides real-time dynamic visualisation. ICG instillation over bilateral inguinal nodes under ultrasound guidance is safe, efficient and has high accuracy than first web space instillation technique in identifying TD and avoiding any inadvertent injuries during oesophageal mobilisation. It is an effective tool for the surgeons planning to negotiate their initial learning curve in robotic-assisted oesophagectomy procedures.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.